- Email: [email protected]
Glaucoma after Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity
Glaucoma after Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity Eric Nudleman, MD, PhD,1,2 Ilkay Kilic Muftuoglu, MD,1 Raouf Gaber, MD,1,4 Joshua Robinson, MD,2,3 Kimberly Drenser, MD, PhD,2 Antonio Capone, MD,2 Michael T. Trese, MD2 Purpose: To report the incidence of, and factors related to, glaucoma after lens-sparing vitrectomy (LSV) surgery in advanced retinopathy of prematurity (ROP). Design: Retrospective case series at a single tertiary referral pediatric vitreoretinal practice. Participants: Four hundred and one eyes from 270 patients were included. Methods: The medical records of patients who underwent LSV for stage 4A, 4B, and 5 ROP were retrospectively reviewed. Data were collected from patient charts including gender, gestational age at birth, birthweight, stage of ROP at presentation, prior treatment (laser or cryotherapy), subsequent retinal surgeries, presence of glaucoma, time to glaucoma (interval between LSV and the onset of glaucoma), date of lensectomy (if performed), and retinal attachment status at last visit. Lensectomy was considered as a time-dependent covariate in the analysis. Main Outcome Measures: Incidence of glaucoma and potential risk factors for time to glaucoma. Results: Among 401 eyes with advanced ROP, 40 eyes (10.0%) had glaucoma during a mean of 3.064.11 years of follow-up. The incidence of glaucoma was 6.9% (17/247) in stage 4A, 12.0% (16/133) in stage 4B, and 33.3% (7/21) in stage 5 ROP. Twenty-one percent of eyes (87/401) required lensectomy at a mean of 1.232.19 years after LSV. In univariate analysis, having stage 5 ROP (vs. stage 4 ROP) and presence of lensectomy were found to be significantly associated with time to glaucoma (hazard ratio ¼ 6.76, 95% confidence interval ¼ 2.19e20.88, P ¼ 0.001; hazard ratio ¼ 3.06, 95% confidence interval ¼ 1.56e6.0, P ¼ 0.001, respectively). In multivariate analysis, lensectomy was the only significant independent factor associated with time to glaucoma (hazard ratio ¼ 2.76, 95% confidence interval ¼ 1.371e5.581, P ¼ 0.004). Conclusions: Patients with more severe ROP had a higher incidence of glaucoma after lens-sparing vitrectomy. If a patient required lensectomy owing to progression of ROP and/or presence of lens opacity, then the hazard of having glaucoma significantly increased compared with those without lensectomy. Ophthalmology 2017;-:1e5 ª 2017 by the American Academy of Ophthalmology
Retinopathy of prematurity (ROP) is 1 of the leading causes of childhood blindness worldwide, accounting for 6% to 18% of all cases.1 Although laser photocoagulation significantly reduces the severity of vision loss in patients with threshold ROP,2,3 among the cases of eyes undergoing laser treatment, approximately 10% of children require vitrectomy for the treatment of ROP-associated retinal detachment (Ciaccia S, Ibarra M, Capone A, Trese M. Fiveyear incidence of blindness from retinopathy of prematurity. Presented at: ARVO Annual Meeting, 2004: Fort Lauderdale, FL). Lens-sparing vitrectomy (LSV) has been shown to be an effective surgical approach with a relatively high success rate, ranging from 42.6% to 82.1% depending on the severity.4e6 Despite successful retina reattachment, approximately 2100 infants in the United States are affected annually by long-term complications of ROP, such as corneal opacity, glaucoma, and amblyopia, which limits the visual recovery.7e10 Childhood glaucoma is a severe condition resulting from various pathologies including congenital defects, trauma, inflammation, or secondary to ocular diseases, with an ª 2017 by the American Academy of Ophthalmology Published by Elsevier Inc.
incidence of 1/10 000 to 1/30 000 live births in the western countries.11 Although there have been several reports suggesting that the risk of glaucoma increases after vitrectomy in adults,12,13 there is limited data on the risk of glaucoma after LSV. Here we report the incidence of, and risk factors for, glaucoma in a large series of eyes that underwent LSV for advanced ROP.
Methods The charts of 279 patients who underwent vitrectomy for stage 4A, 4B, and 5 ROP between 1992 and 2013 at a single tertiary referral pediatric retina clinic were retrospectively reviewed. Institutional review board approval for the data collection and the study was granted by the Western Institutional Review Board. Baseline characteristics of the patients, including gender, gestational age at birth, birthweight, stage of ROP at presentation, and prior ablative treatment (laser photocoagulation or cryotherapy), were collected from the charts. Date of vitrectomy, subsequent retinal surgeries (if performed), status of the retina at last follow-up, history of lensectomy, date of lensectomy
https://doi.org/10.1016/j.ophtha.2017.11.009 ISSN 0161-6420/17
1
Ophthalmology Volume -, Number -, Month 2017 Table 1. Baseline Demographics of the Patients who Underwent Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity
Eyes, n Patients, n Gender, male, % Mean gestational age SD, weeks (range) Mean birth weight SD, grams (range) Median follow-up duration, years (interquartile range) Mean follow-up duration SD, years (range)
Stage 4A
Stage 4B
Stage 5
Total
247 172 58.1 25.31.8 (21.7e32.0)
133 80 51.2 25.42.2 (22.7e33)
21 18 55.5 25.22.0 (22.71e30.0)
401 270 55.9 25.41.9 (21.71e33.0)
768239 (420e1758)
764258.9 (400e1700)
761.7288.2 (350e1516)
776249.72 (400e1758)
0.4 (0.1e3.7)
2.2 (0.3e7.7)
4.11 (2.8e7.1)
0.9 (0.1e5.2)
2.33.5 (0.0e16.7)
4.24.7 (0.0e20.1)
5.04.0 (0.1e14.0)
3.14.1 (0.0e20.1)
SD ¼ standard deviation.
(if performed), and presence of glaucoma were recorded. Patients with a history of retinal surgery at an outside institution before presentation and eyes with scleral buckles (prior, or at time of vitrectomy) were excluded. Glaucoma was defined as an intraocular pressure (IOP) measurement during at least 2 consecutive examinations under anesthesia of 23 mmHg. This number was chosen as a cutoff because it is at least twice the average IOP of normal children aged 0 to 1 year.14e16 LSV was performed as previously described.17,18 The mean follow-up duration was considered as the interval between the date of LSV and the last follow-up visit that included IOP measurement.
Statistical Analysis Descriptive continuous variables were presented using the mean and standard deviation, whereas categorical variables were reported as proportions (%). Kaplan-Meier survival plots were generated to show time to glaucoma in each stage of ROP and in the overall study population. Survival time before development of glaucoma was considered to be the interval between LSV and onset of glaucoma. Frailty models, which are a multivariate randomeffects survival model, were used to model the association between the hazard of developing glaucoma and individual baseline variables including gender, birthweight, gestational age at birth, presence of prior ablative treatment, and stage of ROP. The proportions of categorical variables were compared using a mixed model with a subject random effect to account for the inter-eye
correlation of the same subjects. For the predictors of glaucoma, frailty models were preferred to Cox regression modeling so as to account for the fact that subjects contribute 2 eyes to the study, which are more likely to share similar characteristics to each other than to other patient eyes, violating the assumption that each eye is independent. Because some patients had lensectomy at some point after LSV owing to progression of ROP and/or presence of lens opacity, lensectomy was included as a time-dependent covariate in this analysis. Each covariate was initially fitted in a univariate analysis and significant predictors of survival before developing glaucoma (P < 0.05) were used and re-evaluated in a multivariate analysis. Statistical analysis was carried out using SPSS version 23.0 (SPSS, Inc, Chicago, IL) and R software (version 3.1-122; R Foundation for Statistical Computing, Vienna, Austria). For all hypothesis tests, 2-sided P value of <0.05 was considered as statistically significant.
Results Among 419 eyes of 279 patients, 401 eyes of 270 patients met inclusion criteria. The baseline demographics of the patients are summarized in Table 1. Most of the eyes (61.6%) had stage 4A disease, 133 eyes (33%) had stage 4B, and 21 eyes (5.2%) had stage 5 ROP. The median follow-up duration was 0.85 years in the study population (interquartile range: 0.12e5.21 years).
Table 2. Clinical Characteristics of the Eyes That Underwent Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity Severity of ROP
Number of eyes Prior ablative treatment, present, n (%) PC, n (%) Cryo, n (%) PCþCryo, n (%) Presence of glaucoma, n Need for lensectomy during follow-up, n (%) Mean interval between first LSV and lensectomy SD, years (range) Subsequent retinal surgery, n (%) Attached retina at last follow-up, n (%)
Stage 4A
Stage 4B
Stage 5
Total
247 244 (98.7%) 242 (97.9%) 2 (0.8%) 0 17 (6.9%) 28 (11.3%) 1.40.3 (0.1e13.5)
133 125 (93.9%) 124 (93.2%) 0 1 (0.7%) 16 (12.0%) 44 (33.0%) 1.11.9 (0.1e9.0)
21 20 (95.2%) 19 (90.5%) 0 1 (4.8%) 7 (33.3%) 15 (71.4%) 1.31.7 (0.0e5.9)
401 389 (97.0%) 385 (96.0%) 2 (0.5%) 2 (0.5%) 40 (10.0%) 87 (21.7%) 1.22.2 (0.0e13.5)
38 (15.4%) 237 (96%)
41 (30.8%) 121 (91%)
7 (33.3%) 17 (81%)
86 (21.4%) 375 (93.5%)
Cryo ¼ cryotherapy; LSV ¼ lens-sparing vitrectomy; PC ¼ laser photocoagulation; ROP ¼ retinopathy of prematurity; SD ¼ standard deviation.
2
Nudleman et al
Glaucoma after Lens-Sparing Vitrectomy anterior synechaie. Of the 40 eyes with glaucoma, 55.0% (22 eyes) underwent lensectomy at a mean of 1.451.9 years (range: 0.06e6.04 years) after LSV. The lensectomy rate was 18.2% (66/ 361) in eyes without glaucoma, significantly lower than that of eyes with glaucoma (P < 0.001, chi-square test). The rate of attached retina (95.6%) at last follow-up was statistically higher in eyes without glaucoma than in those with glaucoma (77.5%; P < 0.001, chi-square test). The Kaplan-Meier curves illustrate glaucoma-free survival in the overall study population (Fig 1A) and in different stages of advanced ROP (Fig 1B). The estimated cumulative proportion of being glaucoma-free was 92.0% at 5 years, 69.4% at 10 years, and 28% at 15 years follow-up, respectively. Univariate analysis showed that gestational age (P ¼ 0.36), birthweight (P ¼ 0.27), male gender (P ¼ 0.59), and presence of prior ablative treatment including laser photocoagulation and/or cryotherapy (P ¼ 0.83) were not associated with time to glaucoma (Table 3). After lensectomy was considered as a time-dependent variable, presence of lensectomy and having stage 5 ROP (vs. stage 4 ROP) were significantly associated with time of glaucoma (hazard ratio ¼ 6.76, 95% confidence interval ¼ 2.19e20.88, P ¼ 0.001; hazard ratio ¼ 3.06, 95% confidence interval ¼ 1.56e6.0, P ¼ 0.001, respectively). In the multivariate analysis, presence of lensectomy was the only significant independent factor for time to glaucoma (hazard ratio ¼ 2.76, 95% confidence interval ¼ 1.37e5.58, P ¼ 0.004), indicating that patients undergoing lensectomy were 2.76 times more likely to develop glaucoma than those without lensectomy.
Discussion Figure 1. A, Kaplan-Meier plot demonstrating proportion of glaucomafree survival over time in all patients who underwent lens-sparing vitrectomy (LSV) for advanced retinopathy of prematurity (ROP). The numbers in the parentheses indicate the number of eyes at risk at the given followup. B, Kaplan-Meier plot divided by stages of advanced ROP at the time of initial LSV.
All eyes included underwent LSV for advanced ROP. The clinical characteristics of the eyes are shown in Table 2. Following the first LSV, 21% of eyes (n ¼ 86 eyes) required subsequent retinal surgery. Of the subsequent retinal surgeries performed, 6 were second LSV for eyes with stage 4A ROP. At last followup, the retina was attached in 93.5% of eyes (n ¼ 375 eyes). The retina reattachment rate was statistically higher in eyes with stage 4A (96%), and stage 4B (91%) compared with those with stage 5 (81%) disease (P ¼ 0.003). During follow-up, 21.7% (88/ 401) of eyes underwent lensectomy at a mean of 1.232.19 years (range, 0.02e13.5 years) after LSV owing to lens opacity, progression of ROP, or both; the need for lensectomy was highly dependent on disease severity (P < 0.001). Among 401 eyes with advanced ROP, 40 eyes (10.0%) had glaucoma during follow-up. The median time for having glaucoma was 6.63 years (interquartile range: 3.15e10.24 years). The incidence of glaucoma increased with the severity of the disease, with 6.9% (17/247) in stage 4A, 12.0% (16/133) in stage 4B, and 33.3% (7/21) in stage 5 ROP. Importantly, all eyes in this series underwent an initial LSV and had formed anterior chambers without evidence of peripheral
In the study, we demonstrated that incidence of glaucoma after LSV was 10.0% in the overall study population with advanced ROP during a mean of 3.084.06 years of followup, ranging from 6.9% to 33.3% based on the severity of ROP. Although hazard of time to glaucoma increased by 6.76 in eyes with stage 5 ROP compared with those with Table 3. Univariate Analysis for Having Glaucoma at Last Follow-up in Eyes Undergoing Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity
Continuous variables Gestational age Birthweight Categorical variables Gender, male Stage of ROP 5 vs. 4A 4B vs. 4A Lensectomy Prior ablative treatment (PC/Cryo) (yes/no)
Hazard Ratio*
95% CI
P
0.89 0.99
0.69e1.03 0.99e1.0
0.36 0.27
0.84
0.45e1.57
0.59
6.76 1.85 3.06 0.55
2.19e20.88 0.84e4.07 1.56e6.00 0.10e2.87
0.001 0.12 0.001 0.83
CI ¼ confidence interval; Cryo ¼ cryotherapy; PC ¼ laser photocoagulation; ROP ¼ retinopathy of prematurity. *Hazard ratio for gestational age corresponds to a 1-week increase in gestational age. Hazard ratio for birthweight corresponds to a 1-gram increase in birthweight.
3
Ophthalmology Volume -, Number -, Month 2017 stage 4A disease, lensectomy was the most significant independent factor for time to glaucoma, indicating that patients undergoing lensectomy were 2.76 times more likely to develop glaucoma than those without lensectomy. In a previous report,9 the incidence of glaucoma was found to be 33.3% (7/21 eyes) after LSV in patients presenting with stage 4B and 5 ROP during 5.6 years of follow-up. Interestingly, none of the eyes with attached retina developed glaucoma following surgery. In our study, though retina status was not found to be significant for glaucoma, the percentage of patients with attached retina was higher in eyes without glaucoma. The association between vitrectomy and development of glaucoma has been previously evaluated in adults at early and late postoperative periods.12,13 There are various reasons for the acute postoperative IOP elevation after vitrectomy, including use of intraoperative tamponade, postoperative inflammation, and hemorrhage.19,20 Although immediate IOP rise can be managed with topical medications, up to 15% to 20% of eyes are at risk for delayed-onset glaucoma after pars plana vitrectomy. This has been hypothesized to be attributable to increased diffusion of oxygen, resulting in degeneration of trabecular meshwork cells.13 Open-angle glaucoma is a common complication occurring after removal of childhood cataract, with an incidence ranging from 6% to 58.7%, depending on the follow-up duration and diagnostic criteria.21,22 Similarly, in our study, lensectomy was found to be a significant independent risk factor for glaucoma, indicating that if a patient required lensectomy, it was the most significant variable for time to glaucoma. In a recent prospective study with a set of standard diagnostic criteria for glaucoma, the estimated risk of glaucoma was reported as 17% at 4.8 years after cataract removal, and younger age was found to be the only independent risk factor for glaucoma.23 There has been evidence suggesting that an angle-closure mechanism occurs at a higher incidence in ROP patients.24 Hartnett et al25 prospectively examined the anterior segment of 27 eyes of 17 premature infants with stage 4 and 5 ROP to identify and classify structural characteristics that could predispose the premature infant with ROP to glaucoma. They noted some anatomic changes that may have a developmental origin, such as corneal dystrophy, Barkanlike membrane, and large tunica vasculosaelike vessels, and suggested that abnormal development of the iridocorneal angle and inadequate outflow may contribute to the pathophysiology of glaucoma seen in eyes with advanced ROP. There are some limitations of this study, including high variability in the follow-up duration and lack of optic nerve photographs and functional tests for glaucoma. Because these surgeries were performed at a referral center, some follow-up care was performed by local retina specialists, which may bias the incidence of glaucoma in our series. Despite these limitations, here we report the incidence of glaucoma and related risk factors for developing glaucoma in a large series of patients who underwent LSV for advanced ROP. In summary, the incidence of glaucoma, which increases with the severity of ROP, is 10.0% in all study populations undergoing LSV for advanced ROP during a mean of 3
4
years of follow-up. Twenty-one percent of eyes required lensectomy after LSV during follow-up, and the presence of lensectomy was the only independent factor for time to glaucoma. These results imply that these complicated children should be closely monitored, particularly if lensectomy is performed, for long-term complications to facilitate a better visual prognosis.
References 1. Chiang MF, Melia M, Buffenn AN, et al. Detection of clinically significant retinopathy of prematurity using wide-angle digital retinal photography: a report by the American Academy of Ophthalmology. Ophthalmology. 2012;119:1272e1280. 2. Repka MX, Tung B, Good WV, et al. Outcome of eyes developing retinal detachment during the Early Treatment for Retinopathy of Prematurity study. Arch Ophthalmol. 2011;129:1175e1179. 3. Early Treatment for Retinopathy of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity. Results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol. 2003;121: 1684e1696. 4. Capone Jr A, Trese MT. Lens-sparing vitreous surgery for tractional stage 4A retinopathy of prematurity retinal detachments. Ophthalmology. 2001;108:2068e2070. 5. El Rayes EN, Vinekar A, Capone A. Three-year anatomic and visual outcomes after vitrectomy for stage 4B retinopathy of prematurity. Retina. 2008;28:568e572. 6. Nudleman E, Robinson J, Rao P, et al. Long-term outcomes on lens clarity after lens-sparing vitrectomy for retinopathy of prematurity. Ophthalmology. 2015;122:755e759. 7. Choi J, Kim JH, Kim SJ, Yu YS. Long-term results of lenssparing vitrectomy for stages 4B and 5 retinopathy of prematurity. Korean J Ophthalmol. 2011;25:305e310. 8. Lad EM, Nguyen TC, Morton JM, Moshfeghi DM. Retinopathy of prematurity in the United States. Br J Ophthalmol. 2008;92:320e325. 9. Hartnett ME, Gilbert MM, Hirose T, et al. Glaucoma as a cause of poor vision in severe retinopathy of prematurity. Graefes Arch Clin Exp Ophthalmol. 1993;231:433e438. 10. Bremer DL, Rogers DL, Good WV, et al. Glaucoma in the Early Treatment for Retinopathy of Prematurity (ETROP) study. J AAPOS. 2012;16:449e452. 11. Steinkuller PG, Du L, Gilbert C, et al. Childhood blindness. J AAPOS. 1999;3:26e32. 12. Chang S. LXII Edward Jackson lecture: open angle glaucoma after vitrectomy. Am J Ophthalmol. 2006;141:1033e1043. 13. Koreen L, Yoshida N, Escariao P, et al. Incidence of, risk factors for, and combined mechanism of late-onset open-angle glaucoma after vitrectomy. Retina. 2012;32:160e167. 14. Pensiero S, Da Pozzo S, Perissutti P, et al. Normal intraocular pressure in children. J Pediatr Ophthalmol Strabismus. 1992;29:79e84. 15. Erginturk Acar D, Acar U, Ozdemir O, Tunay ZO. Determination of normal values of intraocular pressure and central corneal thickness in healthy premature infantsda prospective longitudinal study. J AAPOS. 2016;20:239e242. 16. Sihota R, Tuli D, Dada T, et al. Distribution and determinants of intraocular pressure in a normal pediatric population. J Pediatr Ophthalmol Strabismus. 2006;43:14e18. 17. Maguire AM, Trese MT. Lens-sparing vitreoretinal surgery in infants. Arch Ophthalmol. 1992;110:284e286.
Nudleman et al
Glaucoma after Lens-Sparing Vitrectomy
18. Maguire AM, Trese MT. Visual results of lens-sparing vitreoretinal surgery in infants. J Pediatr Ophthalmol Strabismus. 1993;30:28e32. 19. Thompson JT, Sjaarda RN, Glaser BM, Murphy RP. Increased intraocular pressure after macular hole surgery. Am J Ophthalmol. 1996;121:615e622. 20. Chen CJ. Glaucoma after macular hole surgery. Ophthalmology. 1998;105:94e100. 21. Mataftsi A, Haidich AB, Kokkali S, et al. Postoperative glaucoma following infantile cataract surgery: an individual patient data meta-analysis. JAMA Ophthalmol. 2014;132:1059e1067. 22. Egbert JE, Wright MM, Dahlhauser KF, et al. A prospective study of ocular hypertension and glaucoma
after pediatric cataract surgery. Ophthalmology. 1995;102: 1098e1101. 23. Freedman SF, Lynn MJ, Beck AD, et al; Infant Aphakia Treatment Study Group. Glaucoma-related adverse events in the first 5 years after unilateral cataract removal in the Infant Aphakia Treatment Study. JAMA Ophthalmol. 2015;133:907e914. 24. Michael AJ, Pesin SR, Katz LJ, Tasman WS. Management of late onset angle-closure glaucoma associated with retinopathy of prematurity. Ophthalmology. 1991;98:1093e1098. 25. Hartnett ME, Gilbert MM, Richardson TM, et al. Anterior segment evaluation of infants with retinopathy of prematurity. Ophthalmology. 1990;97:122e130.
Footnotes and Financial Disclosures Originally received: April 8, 2017. Final revision: October 16, 2017. Accepted: November 6, 2017. Available online: ---.
parental consent was not obtained due to the retrospective nature of this study. Author Contributions: Manuscript no. 2017-801.
1
Department of Ophthalmology, Shiley Eye Institute, University of California San Diego, San Diego, California.
Conception and design: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
2
Data collection: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
3
Analysis and interpretation: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
Associated Retinal Consultants, Oakland University William Beaumont School of Medicine, Royal Oak, Michigan. Department of Ophthalmology, Emory Eye Center, Emory University, Atlanta, Georgia.
4
Department of Ophthalmology, Tanta University, Tanta, Egypt.
Financial Disclosure(s): The authors have no proprietary or commercial interests in any materials discussed in this article. HUMAN SUBJECTS: Human subjects were part of this study protocol. Institutional Review Board (IRB) approval for the data collection and the study was granted by the Western Institutional Review Board. The study adhered to the tenets of the Declaration of Helsinki and complied with the United States Health Insurance Portability and Accountability Act. Patient/
Obtained funding: Not applicable Overall responsibility: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese Abbreviations and Acronyms: IOP ¼ intraocular pressure; LSV ROP ¼ retinopathy of prematurity.
¼
lens-sparing
vitrectomy;
Correspondence: Michael T. Trese, MD, Associated Retinal Consultants, William Beaumont Hospital, 3535 West Thirteen Mile Road, Suite 344, Royal Oak, MI 48073. E-mail: [email protected].
5
Retinopathy of prematurity (ROP) is 1 of the leading causes of childhood blindness worldwide, accounting for 6% to 18% of all cases.1 Although laser photocoagulation significantly reduces the severity of vision loss in patients with threshold ROP,2,3 among the cases of eyes undergoing laser treatment, approximately 10% of children require vitrectomy for the treatment of ROP-associated retinal detachment (Ciaccia S, Ibarra M, Capone A, Trese M. Fiveyear incidence of blindness from retinopathy of prematurity. Presented at: ARVO Annual Meeting, 2004: Fort Lauderdale, FL). Lens-sparing vitrectomy (LSV) has been shown to be an effective surgical approach with a relatively high success rate, ranging from 42.6% to 82.1% depending on the severity.4e6 Despite successful retina reattachment, approximately 2100 infants in the United States are affected annually by long-term complications of ROP, such as corneal opacity, glaucoma, and amblyopia, which limits the visual recovery.7e10 Childhood glaucoma is a severe condition resulting from various pathologies including congenital defects, trauma, inflammation, or secondary to ocular diseases, with an ª 2017 by the American Academy of Ophthalmology Published by Elsevier Inc.
incidence of 1/10 000 to 1/30 000 live births in the western countries.11 Although there have been several reports suggesting that the risk of glaucoma increases after vitrectomy in adults,12,13 there is limited data on the risk of glaucoma after LSV. Here we report the incidence of, and risk factors for, glaucoma in a large series of eyes that underwent LSV for advanced ROP.
Methods The charts of 279 patients who underwent vitrectomy for stage 4A, 4B, and 5 ROP between 1992 and 2013 at a single tertiary referral pediatric retina clinic were retrospectively reviewed. Institutional review board approval for the data collection and the study was granted by the Western Institutional Review Board. Baseline characteristics of the patients, including gender, gestational age at birth, birthweight, stage of ROP at presentation, and prior ablative treatment (laser photocoagulation or cryotherapy), were collected from the charts. Date of vitrectomy, subsequent retinal surgeries (if performed), status of the retina at last follow-up, history of lensectomy, date of lensectomy
https://doi.org/10.1016/j.ophtha.2017.11.009 ISSN 0161-6420/17
1
Ophthalmology Volume -, Number -, Month 2017 Table 1. Baseline Demographics of the Patients who Underwent Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity
Eyes, n Patients, n Gender, male, % Mean gestational age SD, weeks (range) Mean birth weight SD, grams (range) Median follow-up duration, years (interquartile range) Mean follow-up duration SD, years (range)
Stage 4A
Stage 4B
Stage 5
Total
247 172 58.1 25.31.8 (21.7e32.0)
133 80 51.2 25.42.2 (22.7e33)
21 18 55.5 25.22.0 (22.71e30.0)
401 270 55.9 25.41.9 (21.71e33.0)
768239 (420e1758)
764258.9 (400e1700)
761.7288.2 (350e1516)
776249.72 (400e1758)
0.4 (0.1e3.7)
2.2 (0.3e7.7)
4.11 (2.8e7.1)
0.9 (0.1e5.2)
2.33.5 (0.0e16.7)
4.24.7 (0.0e20.1)
5.04.0 (0.1e14.0)
3.14.1 (0.0e20.1)
SD ¼ standard deviation.
(if performed), and presence of glaucoma were recorded. Patients with a history of retinal surgery at an outside institution before presentation and eyes with scleral buckles (prior, or at time of vitrectomy) were excluded. Glaucoma was defined as an intraocular pressure (IOP) measurement during at least 2 consecutive examinations under anesthesia of 23 mmHg. This number was chosen as a cutoff because it is at least twice the average IOP of normal children aged 0 to 1 year.14e16 LSV was performed as previously described.17,18 The mean follow-up duration was considered as the interval between the date of LSV and the last follow-up visit that included IOP measurement.
Statistical Analysis Descriptive continuous variables were presented using the mean and standard deviation, whereas categorical variables were reported as proportions (%). Kaplan-Meier survival plots were generated to show time to glaucoma in each stage of ROP and in the overall study population. Survival time before development of glaucoma was considered to be the interval between LSV and onset of glaucoma. Frailty models, which are a multivariate randomeffects survival model, were used to model the association between the hazard of developing glaucoma and individual baseline variables including gender, birthweight, gestational age at birth, presence of prior ablative treatment, and stage of ROP. The proportions of categorical variables were compared using a mixed model with a subject random effect to account for the inter-eye
correlation of the same subjects. For the predictors of glaucoma, frailty models were preferred to Cox regression modeling so as to account for the fact that subjects contribute 2 eyes to the study, which are more likely to share similar characteristics to each other than to other patient eyes, violating the assumption that each eye is independent. Because some patients had lensectomy at some point after LSV owing to progression of ROP and/or presence of lens opacity, lensectomy was included as a time-dependent covariate in this analysis. Each covariate was initially fitted in a univariate analysis and significant predictors of survival before developing glaucoma (P < 0.05) were used and re-evaluated in a multivariate analysis. Statistical analysis was carried out using SPSS version 23.0 (SPSS, Inc, Chicago, IL) and R software (version 3.1-122; R Foundation for Statistical Computing, Vienna, Austria). For all hypothesis tests, 2-sided P value of <0.05 was considered as statistically significant.
Results Among 419 eyes of 279 patients, 401 eyes of 270 patients met inclusion criteria. The baseline demographics of the patients are summarized in Table 1. Most of the eyes (61.6%) had stage 4A disease, 133 eyes (33%) had stage 4B, and 21 eyes (5.2%) had stage 5 ROP. The median follow-up duration was 0.85 years in the study population (interquartile range: 0.12e5.21 years).
Table 2. Clinical Characteristics of the Eyes That Underwent Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity Severity of ROP
Number of eyes Prior ablative treatment, present, n (%) PC, n (%) Cryo, n (%) PCþCryo, n (%) Presence of glaucoma, n Need for lensectomy during follow-up, n (%) Mean interval between first LSV and lensectomy SD, years (range) Subsequent retinal surgery, n (%) Attached retina at last follow-up, n (%)
Stage 4A
Stage 4B
Stage 5
Total
247 244 (98.7%) 242 (97.9%) 2 (0.8%) 0 17 (6.9%) 28 (11.3%) 1.40.3 (0.1e13.5)
133 125 (93.9%) 124 (93.2%) 0 1 (0.7%) 16 (12.0%) 44 (33.0%) 1.11.9 (0.1e9.0)
21 20 (95.2%) 19 (90.5%) 0 1 (4.8%) 7 (33.3%) 15 (71.4%) 1.31.7 (0.0e5.9)
401 389 (97.0%) 385 (96.0%) 2 (0.5%) 2 (0.5%) 40 (10.0%) 87 (21.7%) 1.22.2 (0.0e13.5)
38 (15.4%) 237 (96%)
41 (30.8%) 121 (91%)
7 (33.3%) 17 (81%)
86 (21.4%) 375 (93.5%)
Cryo ¼ cryotherapy; LSV ¼ lens-sparing vitrectomy; PC ¼ laser photocoagulation; ROP ¼ retinopathy of prematurity; SD ¼ standard deviation.
2
Nudleman et al
Glaucoma after Lens-Sparing Vitrectomy anterior synechaie. Of the 40 eyes with glaucoma, 55.0% (22 eyes) underwent lensectomy at a mean of 1.451.9 years (range: 0.06e6.04 years) after LSV. The lensectomy rate was 18.2% (66/ 361) in eyes without glaucoma, significantly lower than that of eyes with glaucoma (P < 0.001, chi-square test). The rate of attached retina (95.6%) at last follow-up was statistically higher in eyes without glaucoma than in those with glaucoma (77.5%; P < 0.001, chi-square test). The Kaplan-Meier curves illustrate glaucoma-free survival in the overall study population (Fig 1A) and in different stages of advanced ROP (Fig 1B). The estimated cumulative proportion of being glaucoma-free was 92.0% at 5 years, 69.4% at 10 years, and 28% at 15 years follow-up, respectively. Univariate analysis showed that gestational age (P ¼ 0.36), birthweight (P ¼ 0.27), male gender (P ¼ 0.59), and presence of prior ablative treatment including laser photocoagulation and/or cryotherapy (P ¼ 0.83) were not associated with time to glaucoma (Table 3). After lensectomy was considered as a time-dependent variable, presence of lensectomy and having stage 5 ROP (vs. stage 4 ROP) were significantly associated with time of glaucoma (hazard ratio ¼ 6.76, 95% confidence interval ¼ 2.19e20.88, P ¼ 0.001; hazard ratio ¼ 3.06, 95% confidence interval ¼ 1.56e6.0, P ¼ 0.001, respectively). In the multivariate analysis, presence of lensectomy was the only significant independent factor for time to glaucoma (hazard ratio ¼ 2.76, 95% confidence interval ¼ 1.37e5.58, P ¼ 0.004), indicating that patients undergoing lensectomy were 2.76 times more likely to develop glaucoma than those without lensectomy.
Discussion Figure 1. A, Kaplan-Meier plot demonstrating proportion of glaucomafree survival over time in all patients who underwent lens-sparing vitrectomy (LSV) for advanced retinopathy of prematurity (ROP). The numbers in the parentheses indicate the number of eyes at risk at the given followup. B, Kaplan-Meier plot divided by stages of advanced ROP at the time of initial LSV.
All eyes included underwent LSV for advanced ROP. The clinical characteristics of the eyes are shown in Table 2. Following the first LSV, 21% of eyes (n ¼ 86 eyes) required subsequent retinal surgery. Of the subsequent retinal surgeries performed, 6 were second LSV for eyes with stage 4A ROP. At last followup, the retina was attached in 93.5% of eyes (n ¼ 375 eyes). The retina reattachment rate was statistically higher in eyes with stage 4A (96%), and stage 4B (91%) compared with those with stage 5 (81%) disease (P ¼ 0.003). During follow-up, 21.7% (88/ 401) of eyes underwent lensectomy at a mean of 1.232.19 years (range, 0.02e13.5 years) after LSV owing to lens opacity, progression of ROP, or both; the need for lensectomy was highly dependent on disease severity (P < 0.001). Among 401 eyes with advanced ROP, 40 eyes (10.0%) had glaucoma during follow-up. The median time for having glaucoma was 6.63 years (interquartile range: 3.15e10.24 years). The incidence of glaucoma increased with the severity of the disease, with 6.9% (17/247) in stage 4A, 12.0% (16/133) in stage 4B, and 33.3% (7/21) in stage 5 ROP. Importantly, all eyes in this series underwent an initial LSV and had formed anterior chambers without evidence of peripheral
In the study, we demonstrated that incidence of glaucoma after LSV was 10.0% in the overall study population with advanced ROP during a mean of 3.084.06 years of followup, ranging from 6.9% to 33.3% based on the severity of ROP. Although hazard of time to glaucoma increased by 6.76 in eyes with stage 5 ROP compared with those with Table 3. Univariate Analysis for Having Glaucoma at Last Follow-up in Eyes Undergoing Lens-Sparing Vitrectomy for Advanced Retinopathy of Prematurity
Continuous variables Gestational age Birthweight Categorical variables Gender, male Stage of ROP 5 vs. 4A 4B vs. 4A Lensectomy Prior ablative treatment (PC/Cryo) (yes/no)
Hazard Ratio*
95% CI
P
0.89 0.99
0.69e1.03 0.99e1.0
0.36 0.27
0.84
0.45e1.57
0.59
6.76 1.85 3.06 0.55
2.19e20.88 0.84e4.07 1.56e6.00 0.10e2.87
0.001 0.12 0.001 0.83
CI ¼ confidence interval; Cryo ¼ cryotherapy; PC ¼ laser photocoagulation; ROP ¼ retinopathy of prematurity. *Hazard ratio for gestational age corresponds to a 1-week increase in gestational age. Hazard ratio for birthweight corresponds to a 1-gram increase in birthweight.
3
Ophthalmology Volume -, Number -, Month 2017 stage 4A disease, lensectomy was the most significant independent factor for time to glaucoma, indicating that patients undergoing lensectomy were 2.76 times more likely to develop glaucoma than those without lensectomy. In a previous report,9 the incidence of glaucoma was found to be 33.3% (7/21 eyes) after LSV in patients presenting with stage 4B and 5 ROP during 5.6 years of follow-up. Interestingly, none of the eyes with attached retina developed glaucoma following surgery. In our study, though retina status was not found to be significant for glaucoma, the percentage of patients with attached retina was higher in eyes without glaucoma. The association between vitrectomy and development of glaucoma has been previously evaluated in adults at early and late postoperative periods.12,13 There are various reasons for the acute postoperative IOP elevation after vitrectomy, including use of intraoperative tamponade, postoperative inflammation, and hemorrhage.19,20 Although immediate IOP rise can be managed with topical medications, up to 15% to 20% of eyes are at risk for delayed-onset glaucoma after pars plana vitrectomy. This has been hypothesized to be attributable to increased diffusion of oxygen, resulting in degeneration of trabecular meshwork cells.13 Open-angle glaucoma is a common complication occurring after removal of childhood cataract, with an incidence ranging from 6% to 58.7%, depending on the follow-up duration and diagnostic criteria.21,22 Similarly, in our study, lensectomy was found to be a significant independent risk factor for glaucoma, indicating that if a patient required lensectomy, it was the most significant variable for time to glaucoma. In a recent prospective study with a set of standard diagnostic criteria for glaucoma, the estimated risk of glaucoma was reported as 17% at 4.8 years after cataract removal, and younger age was found to be the only independent risk factor for glaucoma.23 There has been evidence suggesting that an angle-closure mechanism occurs at a higher incidence in ROP patients.24 Hartnett et al25 prospectively examined the anterior segment of 27 eyes of 17 premature infants with stage 4 and 5 ROP to identify and classify structural characteristics that could predispose the premature infant with ROP to glaucoma. They noted some anatomic changes that may have a developmental origin, such as corneal dystrophy, Barkanlike membrane, and large tunica vasculosaelike vessels, and suggested that abnormal development of the iridocorneal angle and inadequate outflow may contribute to the pathophysiology of glaucoma seen in eyes with advanced ROP. There are some limitations of this study, including high variability in the follow-up duration and lack of optic nerve photographs and functional tests for glaucoma. Because these surgeries were performed at a referral center, some follow-up care was performed by local retina specialists, which may bias the incidence of glaucoma in our series. Despite these limitations, here we report the incidence of glaucoma and related risk factors for developing glaucoma in a large series of patients who underwent LSV for advanced ROP. In summary, the incidence of glaucoma, which increases with the severity of ROP, is 10.0% in all study populations undergoing LSV for advanced ROP during a mean of 3
4
years of follow-up. Twenty-one percent of eyes required lensectomy after LSV during follow-up, and the presence of lensectomy was the only independent factor for time to glaucoma. These results imply that these complicated children should be closely monitored, particularly if lensectomy is performed, for long-term complications to facilitate a better visual prognosis.
References 1. Chiang MF, Melia M, Buffenn AN, et al. Detection of clinically significant retinopathy of prematurity using wide-angle digital retinal photography: a report by the American Academy of Ophthalmology. Ophthalmology. 2012;119:1272e1280. 2. Repka MX, Tung B, Good WV, et al. Outcome of eyes developing retinal detachment during the Early Treatment for Retinopathy of Prematurity study. Arch Ophthalmol. 2011;129:1175e1179. 3. Early Treatment for Retinopathy of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity. Results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol. 2003;121: 1684e1696. 4. Capone Jr A, Trese MT. Lens-sparing vitreous surgery for tractional stage 4A retinopathy of prematurity retinal detachments. Ophthalmology. 2001;108:2068e2070. 5. El Rayes EN, Vinekar A, Capone A. Three-year anatomic and visual outcomes after vitrectomy for stage 4B retinopathy of prematurity. Retina. 2008;28:568e572. 6. Nudleman E, Robinson J, Rao P, et al. Long-term outcomes on lens clarity after lens-sparing vitrectomy for retinopathy of prematurity. Ophthalmology. 2015;122:755e759. 7. Choi J, Kim JH, Kim SJ, Yu YS. Long-term results of lenssparing vitrectomy for stages 4B and 5 retinopathy of prematurity. Korean J Ophthalmol. 2011;25:305e310. 8. Lad EM, Nguyen TC, Morton JM, Moshfeghi DM. Retinopathy of prematurity in the United States. Br J Ophthalmol. 2008;92:320e325. 9. Hartnett ME, Gilbert MM, Hirose T, et al. Glaucoma as a cause of poor vision in severe retinopathy of prematurity. Graefes Arch Clin Exp Ophthalmol. 1993;231:433e438. 10. Bremer DL, Rogers DL, Good WV, et al. Glaucoma in the Early Treatment for Retinopathy of Prematurity (ETROP) study. J AAPOS. 2012;16:449e452. 11. Steinkuller PG, Du L, Gilbert C, et al. Childhood blindness. J AAPOS. 1999;3:26e32. 12. Chang S. LXII Edward Jackson lecture: open angle glaucoma after vitrectomy. Am J Ophthalmol. 2006;141:1033e1043. 13. Koreen L, Yoshida N, Escariao P, et al. Incidence of, risk factors for, and combined mechanism of late-onset open-angle glaucoma after vitrectomy. Retina. 2012;32:160e167. 14. Pensiero S, Da Pozzo S, Perissutti P, et al. Normal intraocular pressure in children. J Pediatr Ophthalmol Strabismus. 1992;29:79e84. 15. Erginturk Acar D, Acar U, Ozdemir O, Tunay ZO. Determination of normal values of intraocular pressure and central corneal thickness in healthy premature infantsda prospective longitudinal study. J AAPOS. 2016;20:239e242. 16. Sihota R, Tuli D, Dada T, et al. Distribution and determinants of intraocular pressure in a normal pediatric population. J Pediatr Ophthalmol Strabismus. 2006;43:14e18. 17. Maguire AM, Trese MT. Lens-sparing vitreoretinal surgery in infants. Arch Ophthalmol. 1992;110:284e286.
Nudleman et al
Glaucoma after Lens-Sparing Vitrectomy
18. Maguire AM, Trese MT. Visual results of lens-sparing vitreoretinal surgery in infants. J Pediatr Ophthalmol Strabismus. 1993;30:28e32. 19. Thompson JT, Sjaarda RN, Glaser BM, Murphy RP. Increased intraocular pressure after macular hole surgery. Am J Ophthalmol. 1996;121:615e622. 20. Chen CJ. Glaucoma after macular hole surgery. Ophthalmology. 1998;105:94e100. 21. Mataftsi A, Haidich AB, Kokkali S, et al. Postoperative glaucoma following infantile cataract surgery: an individual patient data meta-analysis. JAMA Ophthalmol. 2014;132:1059e1067. 22. Egbert JE, Wright MM, Dahlhauser KF, et al. A prospective study of ocular hypertension and glaucoma
after pediatric cataract surgery. Ophthalmology. 1995;102: 1098e1101. 23. Freedman SF, Lynn MJ, Beck AD, et al; Infant Aphakia Treatment Study Group. Glaucoma-related adverse events in the first 5 years after unilateral cataract removal in the Infant Aphakia Treatment Study. JAMA Ophthalmol. 2015;133:907e914. 24. Michael AJ, Pesin SR, Katz LJ, Tasman WS. Management of late onset angle-closure glaucoma associated with retinopathy of prematurity. Ophthalmology. 1991;98:1093e1098. 25. Hartnett ME, Gilbert MM, Richardson TM, et al. Anterior segment evaluation of infants with retinopathy of prematurity. Ophthalmology. 1990;97:122e130.
Footnotes and Financial Disclosures Originally received: April 8, 2017. Final revision: October 16, 2017. Accepted: November 6, 2017. Available online: ---.
parental consent was not obtained due to the retrospective nature of this study. Author Contributions: Manuscript no. 2017-801.
1
Department of Ophthalmology, Shiley Eye Institute, University of California San Diego, San Diego, California.
Conception and design: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
2
Data collection: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
3
Analysis and interpretation: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese
Associated Retinal Consultants, Oakland University William Beaumont School of Medicine, Royal Oak, Michigan. Department of Ophthalmology, Emory Eye Center, Emory University, Atlanta, Georgia.
4
Department of Ophthalmology, Tanta University, Tanta, Egypt.
Financial Disclosure(s): The authors have no proprietary or commercial interests in any materials discussed in this article. HUMAN SUBJECTS: Human subjects were part of this study protocol. Institutional Review Board (IRB) approval for the data collection and the study was granted by the Western Institutional Review Board. The study adhered to the tenets of the Declaration of Helsinki and complied with the United States Health Insurance Portability and Accountability Act. Patient/
Obtained funding: Not applicable Overall responsibility: Nudleman, Muftuoglu, Gaber, Robinson, Drenser, Capone, Trese Abbreviations and Acronyms: IOP ¼ intraocular pressure; LSV ROP ¼ retinopathy of prematurity.
¼
lens-sparing
vitrectomy;
Correspondence: Michael T. Trese, MD, Associated Retinal Consultants, William Beaumont Hospital, 3535 West Thirteen Mile Road, Suite 344, Royal Oak, MI 48073. E-mail: [email protected].
5